POINT: Census citizenship question will result in misrepresentation

The new citizenship question on the 2010 Census seems innocuous. Why not find out how many citizens are in the United States?

But the Census question does not appear in a vacuum. Instead, in a politically charged environment regarding immigrants — distinctly marked by President Donald Trump’s promises to build a wall between the United States and Mexico and his pledge to ban immigration of people from Muslim-majority countries — it carries a more sinister connotation. At a time when local police departments too often turn immigrants over to Immigration and Customs Enforcement, it’s not difficult to see why immigrants might be scared to share data about their legal status with the government.

And so a likely consequence is clear. Adding an unnecessary citizenship question to the census — which everyone is legally required to complete —  will discourage response rates from immigrant communities, leading to a population undercount, which has important implications for political representation and policy research.

Immigrants and communities of color already have abysmal response rates compared to white citizens on government surveys. The 2010 Census — which did not include a question about citizenship — overcounted white Americans, while undercounting Hispanic Americans, black Americans, American Indian and Alaska Native Americans. That undercount might only become more drastic with this policy change.

Undercounting the population could make a significant difference in the drawing of Congressional districts, shifting political power away from states with a high percentage of noncitizens — often blue states — and toward states with less noncitizens —  often red states. That electoral math, Republicans are hoping, might keep them in power when new lines are drawn in 2020. Such a strategy comes off the heels of Republican gerrymandering in 2010, in which $30 million invested in state legislature battles enabled Republican victories in states where legislatures could draw the lines for Congressional districts. As a result, Republicans secured a 234-201 majority in the House of Representatives, even though Republican candidates received 1.4 million fewer votes than Democratic candidates in 2012. Changing the census would only further such gerrymandering.

Some conservative politicians think that the population counts for Congressional districts shouldn’t include noncitizens at all. Florida Sen. Marco Rubio tweeted last week, “districts apportioned based on # of people not here legally dilutes the political representation of citizens & legal residents.”

It’s true that, when noncitizens are counted for representation, citizens have less proportional political power. But even though noncitizens — including legal permanent residents, noncitizens with other types of legal status and undocumented residents — can’t vote, they are still impacted by the policies implemented by elected officials.

“Nonvoters have an important stake in many policy debates,” Supreme Court Justice Ruth Bader Ginsburg wrote for the majority in Evenwel v. Abbott, which found that states cannot be forced to draw Congressional districts based on voter-eligible populations.

The 2016 American Community Survey estimates that 22 million people in the United States are noncitizens. Moreover, immigrant communities are often concentrated in urban areas with 1.8 million noncitizens in Los Angeles alone. Neglecting noncitizens in our population counts has the effect of ignoring whole communities. Redistricting based on voter-eligible populations would dilute the political representation of legal permanent residents and other immigrants.

Regardless of electoral math, an undercount of the national population could seriously harm the disbursement of federal funds. Researchers at George Washington Institute of Public Policy found that approximately 300 federal programs administer $800 billion to states based on Census population counts. And for each person missing in the 2010 Census, 37 states lost a median of $1,091 in the fiscal year 2015.

By increasing the likelihood of an undercount, the added question reduces the census’ reliability as a scientific data source for policymakers to consult. That could have real implications for social science and health research, which tomorrow’s policy is contingent upon. Moreover, there isn’t a pressing research need to ask about citizenship in the census; the American Community Survey already estimates citizenship data. The major difference is that population counts based on the American Community Survey don’t form the basis for drawing Congressional districts.

Adding the citizenship question might reduce response rates and skew population counts away from immigrants, which is the underlying purpose of the policy change. This insidious political maneuver could result in underrepresentation of immigrants in Congress, and failure to provide necessary resources and accurate data for policy and social science research.

Constituents should start calling Congress: Even though the census seems complicated and inconsequential, it is anything but. And come November, let’s not forget about this.

Sonali Seth is a senior majoring in policy, planning and development. “Point/Counterpoint” runs Wednesdays.